Nanoenergetic Materials: Enhanced Energy Release from Boron by Aluminum Nanoparticle Addition

Abstract

Boron has the highest enthalpy of oxidation per unit mass (and volume) among metals and metalloids and is an excellent candidate as a solid fuel. However, the native oxide present on the surface limits the available energy and rate of its release during oxidation. Here, we report a simple and effective method that removes the oxide in situ during oxidation via an exothermic thermite reaction with aluminum that enriches the particle in B at the expense of Al. B/Al blends with different compositions are optimized using thermogravimetry and differential scanning calorimetry, and the best sample in terms of energy release is characterized by high-resolution transmission electron microscopy, X-ray photoelectron spectroscopy, energy-dispersive spectroscopy, and X-ray diffraction. All compositions release more energy than the individual components, and the blend containing 10% Al by weight outperforms pure B by 40%. The high energy release is due to the synergistic effect of B oxidation and thermite reaction between Al and B₂O₃. We demonstrate the formation of ternary oxide by the oxidation of the B/Al blend that provides porous channels for the oxidation of B, thereby maximizing the contact of the metal and oxidizer. Overall, the results demonstrate the potential of using B/Al blends to improve the energetic performance of B.

1. Introduction

Boron has shown high promise as a fuel additive for propulsion and energetic applications due to its high gravimetric (58 kJ/g) and volumetric (140 kJ/mL) enthalpies of oxidation.¹⁻⁴ Its ignition performance, however, is hindered by the presence of a native oxide on the surface, which melts at relatively low temperatures (450 °C at atmospheric pressure).⁵⁻⁸ The melting of the oxide shell before the solid core clogs the pores leads to particle agglomeration and acts as a diffusion barrier to the incoming oxidizer, thus delaying the boron (B) oxidation.^1,5 Attempts to overcome these limitations include surface functionalization of B by organic compounds,⁹⁻¹⁴ reduction of the oxide, followed by surface passivation using nonthermal plasma processing,¹⁵ and coating with metals to form composites and metal borides by ball milling and high-temperature sintering methods.¹⁶⁻²³ Functionalization with organic compounds results in the reduction of the amount of energy released per unit mass due to the presence of less energetic materials on the B surface. Nonthermal plasma processing has been shown to be successful in enhancing the energetic performance over untreated boron but requires low-pressure equipment that is harder to scale-up. Size reduction of metal particles in the presence of organic solvents using high-energy ball milling has emerged as an effective method. However, it requires multiple chemicals and processing steps prone to introducing contaminants. Postprocessing separation and drying steps render the available methods more complex and time-consuming. High-temperature sintering leads to agglomeration of B powder due to the low melting temperature of B₂O₃ (450 °C). Molten oxide clogs the pores and slows down the oxidation reaction.^15,17,20

In the quest to accelerate the combustion performance of B, metals such as Mg, Al, Zr, Fe, Ti, and Li have been used in conjunction with B.^{1,6,16,19,21,22} The presence of these metals relieves the accumulation of liquid boron oxide films by forming porous ternary oxides,^16,17,19 and their ignition raises the local temperature of the reaction interface, thereby facilitating the more complete oxidation of boron.¹⁹⁻²² Among these metals, aluminum (Al), whose presence on the earth is ubiquitous, has been studied extensively on its own merit, due to its better reactivity, high gravimetric energy density (31 kJ/g), and relatively low melting point.²³ Al can combine with B to form Al borides,²⁴ which show better thermal stability during storage but release less energy during combustion (40 kJ/g) compared to boron (58 kJ/g).^21,22 The chemical bonding between the two elements leads to significant ignition delays, adding a further detriment in applications that require fast energy release.²² These limitations could be overcome if Al and B form a mechanical blend rather than a chemical compound. Studies on B/Al blends are scarce, but the few that are available suggest that the ignition performance of B/Al blends at a weight ratio of 1:1 with micrometer-sized particles is better than that of aluminum diboride.^24,25 This improvement is attributed to the less serious accumulation of liquid B₂O₃ films during the combustion of the blends and the absence of Al–B bonds, making them thermodynamically favorable. Pivkina et al.,²⁶ with the help of thermochemical analysis, established that boron oxide on mixing with Al and Mg powders undergoes thermite reaction at temperatures between 620 and 860 °C to form refractory oxides and active B. Other studies include the investigation of Al–Mg alloys and Mg/B solid solutions.²⁷⁻²⁹ When we add dopants to the metals, careful investigation of segregation dynamics is important. Saidi et al.^27,28 studied the surface segregation dynamics in Al–Mg alloys by developing a robust atomistic potential based on machine learning principles. Interestingly, the Mg concentration of 12–20% in an Al–Mg alloy displays the weakest surface segregation and more uniform distribution of Mg. This model can predict energetically favorable surface positions in metal blends and can be helpful in optimizing the process. Our recent work on Mg/B solid solutions²⁹ also exhibits the potential of more reactive Mg to enhance the oxidation heat release from B particles at lower temperatures. Hence, there is a need to study the effect of Al nanoparticles (NPs) on the energetic performance of B and to optimize the blend to extract the maximum chemical energy at lower temperatures.

Al can reduce boron oxide to produce elemental boron and Al oxide, according to the thermite reaction 1:

1

The reaction is exothermic and frees B from its oxide that may further oxidize and release energy, contributing to the overall energy release. Al effectively acts as a sacrificial element that extracts the energy of B, which is trapped in the form of B₂O₃. We suggest that the reduction of boron oxide by Al can be leveraged to produce B/Al blends with substantially superior performance than B alone. If this hypothesis is correct, optimum performance should be realized with smaller sized Al, as its ignition and melting coincide with the melting of B₂O_3, bringing the reactants in intimate contact and triggering a thermite reaction. Nanosized metal particles have their own advantages in energetic applications. They exhibit fast ignition, enhanced reaction kinetics, and more complete oxidation relative to micrometer-sized particles.^6,7 In this study, we demonstrate the superior performance of B/Al blends with respect to energy release and ignition temperatures and identify the optimum amount of Al that must be added to enhance the oxidation performance and maximize the energy of the blend.

2. Experimental Methods

Boron particles (99.5%, 500 nm, Nanoshel) and Al particles (99.9%, 70 nm, US Research Nanomaterials) were used. Al and B particles were mixed in different weight proportions [Al weight (wt) % of 5, 10, 20, 33, 50, 67, and 80% with the rest B] in glass vials by the magnetic agitation of dry powders using a stirrer and a stir bar to form a homogeneous mixture. This mixing method typically takes 3 min to produce a homogeneous mechanical blend without any organic contamination. The blends were characterized using thermochemical analysis––thermogravimetric analysis (TGA)––and differential scanning calorimetry (DSC) and compared to pure Al and B in terms of oxidation and energy release. Additionally, the samples were characterized by X-ray photoelectron spectroscopy (XPS), high-resolution transmission electron microscopy (HRTEM), high-angle annular dark-field (HAADF)-scanning TEM (STEM)-energy dispersive spectroscopy (EDS), and X-ray diffraction (XRD) analysis.

Thermal analyses––TGA and DSC––were performed on a TA Instruments model Q600 SDT, which provides simultaneous measurements of heat flow and weight change for the samples. The heating rate used was 20 °C/min, and the temperature ranged from 20 to 1000 °C. Analyses were conducted in ultra-zero air at a volumetric flow rate of 100 mL/min for all samples studied. Alumina sample cups (90 μL, TA Instruments) were used for the analyses. XPS was performed using a Physical Electronics VersaProbe II instrument equipped with a monochromatic Al Kα X-ray source and a concentric hemispherical analyzer. A take-off angle of 45° to the surface of the sample was used for all the measurements. High-resolution Al2p spectra were analyzed and quantified to integrate with the Strohmeier equation to calculate the thickness of the surface Al₂O₃. HRTEM and HAADF––STEM––EDS were performed on Talos F200X at 200 kV with an XFEG source and an integrated SuperX EDS system. The particles were dispersed into isopropanol and sonicated for 10 min using a Branson ultrasonicator (model: CPX3800H). Few drops of dispersion were placed on TEM lacey carbon Cu grids (Electron Microscopy Sciences), which were stored for few hours to evaporate the solvent. XRD spectra were obtained using a Malvern Panalytical XPert Pro MPD θ–θ diffractometer with a Cu Kα X-ray source.

3. Results and Discussion

Figure 1 shows the energy measured by DSC during oxidation up to 1000 °C in air. In this temperature range, commercial B releases 30 kJ/g, while Al releases 15 kJ/g of energy. If the two materials oxidized independently of each other, the energy of the mixture would lie on the straight line that connects the unblended components. Interestingly, all blends produce energy that lies above this line, indicating a synergistic chemical interaction between the two materials. Upon addition of Al, the energy increases rapidly above that of commercial B and reaches a maximum at 10 wt % Al. Above 10 wt % Al, the energy decreases gradually to that of pure Al but remains higher than that of B until approximately 40 wt % Al. Under optimum conditions, we obtain the energy of 42 kJ/g, which corresponds to a 40% increase over commercial B particles.

Figure 1.

DSC analysis trends showing energy release from different compositions of B/Al blends. Error bars of ±3% are the result of performing the experiment three times.

Stoichiometric calculations were performed with the support of XPS and HRTEM analyses to determine the amount of Al NPs (70 nm) required to eliminate the native B₂O₃ present on the surface of B (500 nm). Figure 2a,b represents the HRTEM micrographs of B and Al, respectively. The estimated thickness of native oxide on B particles is 6 nm and that of Al NPs is 4.6 nm. High-resolution XPS spectra of B and Al in Figure 2c,d exhibit the composition of native oxides on the surface of B and Al particles. The values of peak intensities of native Al₂O₃ and Al were found from high-resolution XPS analysis (Figure 2d) and used in the Strohmeier equation³⁰ to calculate the thickness of native Al₂O₃ to be 4.6 nm, which agrees with the measured thickness from HRTEM. In this manner, we calculate that the mass of Al needed to fully react with B₂O₃ is 9.6%. This is in very good agreement with the observed maximum at 10 wt % Al in the thermochemical analysis of all blends (Figure 1), in direct support of the hypothesis that the primary effect of Al is to engage in a thermite reaction with boron oxide and not allow its accumulation during oxidation.

Figure 2.

Native oxide thickness on the particles measured using HRTEM: (a) B particles having 6 nm thick oxide, (b) Al NPs having 4.6 nm thick oxide, (c) high-resolution XPS spectra of B 1s, and (d) high-resolution XPS spectra of Al 2p showing the relative composition of metal and oxide on the surface.

We characterize the 10 wt % B/Al blends in further detail and refer for brevity to this sample as BAL10. Figure 3a,b shows the TGA and DSC analyses of BAL10, pure B, and pure Al. The weight gain and energy release due to oxidation are measured and then compared for the three samples. The weight gain is directly proportional to the amount of oxygen bound on the oxide that forms during the heating of the sample in the air. Pure B ignites first and exhibits a sharp weight gain due to oxidation at 550 °C, followed by a slower rate of increase to a net gain of 145% at 1000 °C. The oxidation of Al begins at a somewhat higher temperature (620 °C), and its net weight gain is 46% at 1000 °C, lower than that of B due to the larger molecular weight of Al. If the two metals oxidized independently of each other, the weight gain due to their oxidation would be expected to be somewhere in between the two, that is, at 135%, as indicated in Figure 3a using a black dotted line. Instead, the BAL10 blend follows with a small delay in the profile of B and reaches a higher weight gain of 150% at 1000 °C, 15% higher than the expected value, which clearly confirms the promoting effect of Al on B oxidation, as shown in Figure 3a using a green arrow. The elevated slope of the weight change at higher temperatures (>800 °C) indicates the diminishing effects of liquid B₂O₃ accumulation due to the presence of Al. The DSC profile of the BAL10 blend is distinctly different from that of the pure components (Figure 3b). Pure B releases a peak energy of 30 kJ/g at 650 °C, while Al releases a much sharper peak with less energy (15 kJ/g) at 600 °C. The blend shows two peaks: a sharp peak at 650 °C, followed by a broader peak at 675 °C. The energy release measured from these peaks is 42 kJ/g, which is 40% higher than the energy release of B. It may be noticed that there is a small delay in the ignition of BAL10 as compared to pure B and Al powders. However, the focus of this work is on improving the energetic performance of B, which is achieved in the form of 40% energy enhancement. This is a trade-off between the ignition of the material and its energy release. To identify the origins of exothermic peaks in BAL10, we performed DSC of the blend in an inert argon atmosphere to confirm that the thermite reaction occurred (Figure 4a). We observe a peak (1.2 kJ/g) at 600 °C, which is attributed to the reduction of B₂O₃ by Al. This confirms that the thermite reaction starts first, and the heat generated from the reaction promotes oxidation, causing the more complete reaction of B present in the sample. XRD of the blend (S1 in Supporting Information) shows that its main components are Al, Al₂O₃, B, and B₂O₃. Hence, the possible reaction is between Al and B₂O₃ (reaction 1), as suggested by the thermodynamic calculations. Active B is formed by the reduction of B₂O₃ in the thermite reaction, and its subsequent oxidation makes a significant contribution to the amount of energy released, leading to a broader exothermic peak.

2

Figure 3.

(a) TGA and (b) DSC analyses of Al, B, and BAL10 showing improvements in oxidation and energy release in BAL10 blend as compared to Al NPs and B particles.

Figure 4.

Thermite reaction and product: (a) DSC analysis comparison of BAL10 blend in air and argon showing the exothermic thermite reaction between Al and B₂O₃, (b) STEM–EDS micrograph of the thermite reaction product of BAL10 showing Al₂O₃ formed and B particles with reduced surface oxide, (c) STEM–HAADF image of the thermite reaction product, and (d) STEM–EDS micrograph showing the distribution of B and Al in the resultant particles from the thermite reaction.

Figure 4b–d are showing the representative STEM–EDS micrographs of BAL10 after undergoing a thermite reaction (reaction 1). Due to this reaction, Al is oxidized into Al₂O₃ by reducing B₂O₃ present on the B surface into B. The micrograph in Figure 4b shows a particle having Al and O distributed over one another, which is essentially Al₂O_3, and another particle with oxygen-free B because of a thermite reaction. It should be noted that this micrograph is just the representative image of the product of the thermite reaction. Figure 4c shows the dark-field STEM image of the particles, and Figure 4d reveals the distribution of Al and B without oxygen, which shows a distribution of B (red) on the surface of Al (blue), pointing toward the formation of ternary oxides of B and Al during the thermite reaction. However, the more accurate evidence is demonstrated by the XRD analysis shown in Figure 5e. Based on this information, we attribute the first exothermic peak of the blend to the combined effect of the reduction of B₂O₃ by Al (ΔG° = −859 kJ/mol) plus the oxidation of Al (ΔG° = −1691 kJ/mol) and the second peak to the oxidation of B (ΔG° = −832 kJ/mol). These conclusions are corroborated by the DSC profiles of blends at other compositions, which show a systematic rise in the first peaks and a decrease in the area of the second peaks as more Al is added to the blend (S2 in Supporting Information). The elemental distribution of the oxidation product of sample BAL10 by HAADF–STEM–EDS is shown in the micrographs of Figure 5a–d. B, Al, and O are distributed over one another, indicating the possibility of the formation of ternary oxides during oxidation. Their formation was confirmed by XRD analysis on the same sample, as shown in Figure 5e.

Figure 5.

HAADF–STEM–EDS images showing (a) STEM micrograph and distribution of (b) B, (c) Al, and (d) O. (e) XRD diffractogram of the oxidation product of BAL10.

The higher extent of oxidation and the corresponding heat release from BAL10 (Figure 3) can be attributed to the absence of a liquid B₂O₃ layer due to the formation of a ternary oxide containing Al, B, and O, as observed from the HAADF–STEM–EDS and XRD analyses of the oxidized sample along with the study of the phase diagram of the Al₂O₃/B₂O₃ mixture.³¹ These ternary oxides take care of the limitation of molten B₂O₃ by providing porous channels for oxidizers to trigger B oxidation.¹⁷ The presence of ternary oxides reduces the clogging of B pores and aggregation of particles by molten B₂O₃.^1,17 During the oxidation of BAL10, Al reacts with B₂O₃ to form Al₂O₃ and then B oxides to B₂O₃. In the end, both oxides are present in the oxidation product. Effectively, Al₂O₃ combines with B₂O₃ to form a ternary oxide. Interestingly, Al₂O₃ has a lower heat capacity (0.88 J/g K) than B₂O₃ (1.5 J/g K), so this combination in fact decreases the overall heat capacity of the reaction products and lessens the heat arresting effect of the oxides. In DSC analysis (Figure 3b), Al NPs release 15 kJ/g, and an exothermic peak appears slightly early as compared to B (30 kJ/g) because of the nanosize (70 nm) of the particle. The onset of the BAL10 blend is almost the same as that of Al, but a broader exotherm with a higher energy release is observed because of the occurrence of simultaneous thermite and oxidation reactions between Al, B, and B₂O₃, providing a synergistic effect to get a superior energy release. Thus, the maximum energy release from the BAL10 blend (42 kJ/g) is due to the conversion of native B₂O₃ into an energetic component through an exothermic thermite reaction with Al. This means that we can extract more energy from B at lower temperatures by removing a kinetic barrier (native oxide) without losing energy. Hence, we conclude that the presence of nano-Al in optimum stoichiometric proportion (10 wt %) with B particles causes interfacial exothermic reactions and leads to the formation of nanoenergetic blends with enhanced energy release.

4. Conclusions

In summary, we have developed a simple and highly efficient method to extract higher amounts of chemical energy from B at low temperatures by removing the kinetic barrier using Al. The process can be easily scaled up without causing any chemical contamination and takes only a few minutes, unlike other complex techniques in the literature that take several hours to days. The significant improvement in the energy release from the blends is primarily due to the synergistic effect of exotherms from B oxidation and thermite reaction between Al and B₂O₃ and secondarily due to the formation of porous channels of ternary oxides of Al and B that increases the intimate contact of B and the oxidizer. The B/Al blends can find significant applications in solid fuels and energetic systems used in civilian and military applications due to the superior energy release at lower temperatures.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.2c02691.

• XRD analysis of BAL10 blend and DSC results of other samples with different B/Al compositions (PDF)

The authors declare no competing financial interest.